Teasing Out the Overall Survival Benefit With Adjustment for Treatment Switching to Multiple Treatments

Abstract

In oncology clinical trials, characterizing the long-term overall survival (OS) benefit for an experimental treatment is often unobservable if some patients in the control group switch to drugs in the experimental group and/or other cancer treatments after disease progression. A key question often raised by payers and reimbursement agencies is how to estimate the true benefit of the experimental treatment on OS that would have been estimated if there were no treatment switches. Several commonly used statistical methods are available to estimate OS benefit while adjusting for treatment switching, ranging from naive exclusion or censoring approaches to more advanced methods including inverse probability of censoring weighting (IPCW), two-stage model (2-stage), iterative parameter estimation (IPE) algorithm or rank-preserving structural failure time models (RPSFTM). However, many clinical trials now have patients in the control group switching to different treatment regimens other than the experimental treatments. While some existing methods including RPSFTM and IPE cannot handle complicated scenarios such as multilevel switching, other methods such as IPCW and two-stage methods are restricted with strong assumptions. To address this challenge, we propose two additional methods: stratified RPSFTM and random-forest-based prediction. A simulation study is conducted to assess the performance of the existing methods along with the two newly proposed approaches.